Microplastic pollution in agricultural soil threatens soil quality and human health which has attracted extensive attention worldwide. However, there is no uniform standard for microplastic extraction methods and the identification of microplastic component in different typical agricultural soils. In this study, an artificial simulation adding experiment was used in eight typical agricultural soil samples in China. The aim of the study was to use different methods for extraction, comparing the extraction rates of four microplastics and their influence on polymer identification using ATR- FTIR. The two separate methods were oil extraction (water + oil and saturated NaCl solution + oil), and density method (saturated NaCl solution). The four types of microplastics include polyethylene terephthalate (PET), polypropylene (PP), polystyrene (PS), and polyethylene (PE). We found that the oil extraction method effectively extracted four types of microplastics in agricultural soils, which varied from 83.33% to 100.00%. However, the extraction rate of PET under the oil extraction method and PP under the density method from Southern laterite area was lower than other soils. The presence of iron and aluminum ions influenced the extraction rates of microplastics in the Southwest laterite area. With the increase in microplastic density, the extraction rates of the density method decrease. The oil extraction methods with the cleaning of residual oil were recommended for the higher density microplastics. The density method was recommended for the lighter microplastics in agricultural soils. However, these two extraction methods were not ideal to extract the microplastics from the Southern laterite area and the appropriate extraction methods for laterite need to be further studied in the future. Our results can provide technical support for the extraction treatment and scientific microplastic pollution control of typical agricultural soils with different erosion areas.